A Semi-Vectorial Hybrid Morphological Segmentation of Multicomponent Images Based on Multithreshold Analysis of Multidimensional Compact Histogram

In this work, we propose an original approach of semi-vectorial hybrid morphological segmentation for multicomponent images or multidimensional data by analyzing compact multidimensional histograms based on different orders. Its principle consists first of segment marginally each component of the multicomponent image into different numbers of classes fixed at K. The segmentation of each component of the image uses a scalar segmentation strategy by histogram analysis;we mainly count the methods by searching for peaks or modes of the histogram and those based on a multi-thresholding of the histogram. It is the latter that we have used in this paper, it relies particularly on the multi-thresholding method of OTSU. Then, in the case where i) each component of the image admits exactly K classes, K vector thresholds are constructed by an optimal pairing of which each component of the vector thresholds are those resulting from the marginal segmentations. In addition, the multidimensional compact histogram of the multicomponent image is computed and the attribute tuples or ‘colors’ of the histogram are ordered relative to the threshold vectors to produce (K + 1) intervals in the partial order giving rise to a segmentation of the multidimensional histogram into K classes. The remaining colors of the histogram are assigned to the closest class relative to their center of gravity. ii) In the contrary case, a vectorial spatial matching between the classes of the scalar components of the image is produced to obtain an over-segmentation, then an interclass fusion is performed to obtain a maximum of K classes. Indeed, the relevance of our segmentation method has been highlighted in relation to other methods, such as K-means, using unsupervised and supervised quantitative segmentation evaluation criteria. So the robustness of our method relatively to noise has been tested.

Multi-proxy quantum group signature scheme with threshold shared verification

A multi-proxy quantum group signature scheme with threshold shared verification is proposed. An original signer may authorize a proxy group as his proxy agent. Then only the cooperation of all the signers in the proxy group can generate the proxy signature on behalf of the original signer. In the scheme, any t or more of n receivers can verify the message and any t - 1 or fewer receivers cannot verify the validity of the proxy signature.

Bipartite Threshold Multi-Secret Sharing Scheme Based on Hypersphere

To address the problem that existing bipartite secret sharing scheme is short of dynamic characteristic, and to solve the problem that each participant can only use secret share once, this paper proposed a bipartite (n1+n2, m1+m2)-threshold multi-secret sharing scheme which combined cryptography and hypersphere geometry. In this scheme, we introduced a bivariate function and a coordinate function over finite field Zp to calculate the derived points of secret share, which can reconstruct the shared secrets by producing the intersection point of hypernormal plane and normal line on the hypertangent plane. At the initial stage the secret dealer distributes to each participant a secret share that can be kept secret based on the intractability of discrete logarithm problem and need not be changed with updating the shared secrets.Each cooperative participant only needs to submit a derived point calculated from the secret share without exposing this secret share during the process of reconstructing the shared secret. Analyses indicate that the proposed scheme is not only sound and secure because of hypersphere geometric properties and the difficulty of discrete logarithm problem, but also efficient because of its well dynamic behavior and the invariant secret share. Therefore, this bipartite threshold multi-secret sharing scheme is easy to implement and is applicable in practical settings.

Single-mode low threshold current multi-hole vertical-cavity surface-emitting lasers

A multi-hole vertical-cavity surface-emitting laser （VCSEL） operating in stable single mode with a low threshold current was produced by introducing multi-leaf scallop holes on the top distributed Bragg-refleetor of an oxidation- confined 850 nm VCSEL. The single-mode output power of 2.6 mW, threshold current of 0.6 mA, full width of half maximum lasing spectrum of less than 0.1 nm, side mode suppression ratio of 28.4 dB, and far-field divergence angle of about 10% are obtained. The effects of different hole depths on the optical characteristics are simulated and analysed, including far-field divergence, spectrum and lateral cavity mode. The single-mode performance of this multi-hole device is attributed to the large radiation loss from the inter hole spacing and the scattering loss at the bottom of the holes, particularly for higher order modes.

Multi-Threshold Algorithm Based on Havrda and Charvat Entropy for Edge Detection in Satellite Grayscale Images

Automatic edge detection of an image is considered a type of crucial information that can be extracted by applying detectors with different techniques. It is a main tool in pattern recognition, image segmentation, and scene analysis. This paper introduces an edge-detection algorithm, which generates multi-threshold values. It is based on non-Shannon measures such as Havrda & Charvat’s entropy, which is commonly used in gray level image analysis in many types of images such as satellite grayscale images. The proposed edge detection performance is compared to the previous classic methods, such as Roberts, Prewitt, and Sobel methods. Numerical results underline the robustness of the presented approach and different applications are shown.

Scheduler Algorithm for Multi-Class Switch with Priority Threshold

The requirement for guaranteed Quality of Service (QoS) have become very essential since there are numerous network base application is available such as video conferencing, data streaming, data transfer and many more. This has led to the multi-class switch architecture to cater for the needs for different QoS requirements. The introduction of threshold in multi-class switch to solve the starvation problems in loss sensitive class has increased the mean delay for delay sensitive class. In this research, a new scheduling architecture is introduced to improve mean delay in delay sensitive class when the threshold is active. The proposed architecture has been simulated under uniform and non-uniform traffic to show performance of the switch in terms of mean delay. The results show that the proposed architecture has achieved better performance as compared to Weighted Fair Queueing (WFQ) and Priority Queue (PQ).

基于空中计算CoMAC架构的不同计算场景叠加符号判决算法

本文研究不同场景下基于空中计算的多址信道计算(CoMAC)架构的覆盖符号决策算法。首先,从理论上分析了XOR、ADD、MOD三种场景中加性高斯白噪声(AWGN)多址接入信道下叠加符号的概率密度分布,提出了一种基于先验概率的最优门限判决策略。其次,推导了系统最优门限及对应误码率的理论表达式。最后,通过仿真验证了不同信噪比、传感器节点个数及先验概率对于该门限判决方案的鲁棒性和可靠性的影响。与通信计算相分离的传统方案相比,空中计算判决方案具有更好的检测性能,为多址接入信道下的信号识别提供了新的参考方案。

An Improved SOI CMOS Technology Based Circuit Technique for Effective Reduction of Standby Subthreshold Leakage

Silicon-on-insulator (SOI) CMOS technology is a very attractive option for implementing digital integrated circuits for low power applications. This paper presents migration of standby subthreshold leakage control technique from a bulk CMOS to SOI CMOS technology. An improved SOI CMOS technology based circuit technique for effective reduction of standby subthreshold leakage power dissipation is proposed in this paper. The proposed technique is validated through design and simulation of a one-bit full adder circuit at a temperature of 27℃, supply voltage, VDD of 0.90 V in 120 nm SOI CMOS technology. Existing standby subthreshold leakage control techniques in CMOS bulk technology are compared with the proposed technique in SOI CMOS technology. Both the proposed and existing techniques are also implemented in SOI CMOS technology and compared. Reduction in standby subthreshold leakage power dissipation by reduction factors of 54x and 45x foraone-bit full adder circuit was achieved using our proposed SOI CMOS technology based circuit technique in comparison with existing techniques such as MTCMOS technique and SCCMOS technique respectively in CMOS bulk technology. Dynamic power dissipation was also reduced significantly by using this proposed SOI CMOS technology based circuit technique. Standby subthreshold leakage power dissipation and dynamic power dissipation were also reduced significantly using the proposed circuit technique in comparison with other existing techniques, when all circuit techniques were implemented in SOI CMOS technology. All simulations were performed using Microwindver 3.1 EDA tool.

Weighted Self-Adaptive Threshold Wavelets for Interpolation Point Selection Used in Interconnect MOR

As process technology development,model order reduction( MOR) has been regarded as a useful tool in analysis of on-chip interconnects. We propose a weighted self-adaptive threshold wavelet interpolation MOR method on account of Krylov subspace techniques. The interpolation points are selected by Haar wavelet using weighted self-adaptive threshold methods dynamically. Through the analyses of different types of circuits in very large scale integration( VLSI),the results show that the method proposed in this paper can be more accurate and efficient than Krylov subspace method of multi-shift expansion point using Haar wavelet that are no weighted self-adaptive threshold application in interest frequency range,and more accurate than Krylov subspace method of multi-shift expansion point based on the uniform interpolation point.

多破片冲击引爆屏蔽装药增强效应机理研究

为揭示多破片耦合冲击作用下屏蔽装药的引爆增强效应及机理,基于非均质炸药冲击引爆机理,建立了单破片、双破片同时、双破片时序冲击引爆屏蔽装药的理论模型.结合有限元仿真,获得了破片数量、破片间距、破片作用时序对屏蔽装药冲击引爆效应的影响规律,揭示了多破片耦合引爆增强机理.结果表明:随破片数量增加,破片撞击产生的冲击波叠加效应使引爆阈值速度指数式降低;破片横向间距增加,冲击波衰减导致叠加程度降低,引爆阈值速度上升并趋近于单破片引爆阈值速度;双破片作用屏蔽装药时序不同,炸药初始爆轰波由于固体波前/波后介质状态不同偏离轴线.

基于多源数据与随机森林方法的城市建成区提取——以郑州市为例

基于夜间灯光数据的阈值分割法在城镇建成区提取研究中被广泛应用,但由于夜间灯光数据分辨率低、灯光溢出和阈值分割法无法顾及区域差异等问题,一定程度上影响了该方法的提取精度。以郑州市为例,以LJ1-01与NPP/VIIRS两种夜间灯光影像为主要数据源,结合Landsat8中分辨率遥感影像、网络城市兴趣点(POI)及路网数据,利用随机森林分类方法对郑州市2018年建成区进行提取,参考土地利用数据,对RF分类法与NTL、VANUI、BANUI、PANUI、RANUI指数等阈值法进行对比实验和精度评价,评估基于多源数据的随机森林分类方法在城市建成区提取中的优势。实验表明,RF比阈值法提取的建成区更接近真实建成区且提取精度更高,具有更好适用性;LJ1-01数据提取的效果和精度总体优于NPP/VIIRS数据;在采用RF分类时,各类特征的重要性在不同夜光数据源中表现差异较大。

基于多目标的三支决策边界域求解与分类方法

三支决策将不确定样本划分至边界域进行延迟决策,但需基于损失函数确定阈值,以划分边界域,然而,损失函数通常需要先验知识,具有一定的主观性,因此对边界域划分能力不足。针对这种问题,构建一种多目标三支决策边界域求解方法,从而更好地划分边界域,提升分类性能。采用贝叶斯规则获取样本的条件概率;设定3个目标,包括降低边界域的不确定性、缩小边界域的大小以及减小整个决策区域的错误分类率,通过融入熵权法的TOPSIS(technique for order preference by similarity to an ideal solution)方法求取最优阈值,该方法采用熵权法计算这3个目标所占的权重,得到最优阈值,获得边界域,进行延迟决策;结合不同分类器对边界域进行分类。通过UCI数据集进行对比实验,根据分类准确率和F1值,表明该方法学习到的阈值能合理地划分边界域,建立的模型能取得更好的分类性能。

基于改进雁群算法的Otsu多阈值图像分割

阈值分割是一种被广泛应用的图像分割技术。然而,传统的最大类间方差法(Otsu算法)在多阈值图像分割中面临着计算复杂度高、执行时间长及分割准确性不够等挑战。针对这些现象,提出一种基于改进雁群优化算法(CBLSGSO)的Otsu多阈值图像分割算法,该算法将Cubic混沌映射模型嵌入雁群算法初始化过程中,提高种群的多样性;提出多区域引导式结构,对种群动态切分并设计不同的进化机制,扩大种群寻优范围;引入自适应正余弦策略和蝴蝶算法搜索策略,提高算法的收敛精度,有效地平衡了算法的全局寻优能力和局部寻优能力。为验证改进后Otsu算法性能,选取ACC、Jaccard、Specificity、F1-score、FSIM、SSIM和PSNR等指标作为评价指标,并与近年来不同学者提出的图像分割算法进行实验对比,验证了算法的有效性。实验结果表明,基于改进雁群算法的Otsu图像分割法能更快速精确地解决复杂图像分割问题。

轴承自驱式独立退化轨迹构建与剩余寿命灰色预测

针对轴承退化轨迹的个体差异性及退化阶段划分的人为主观性问题,提出了自监测数据驱动的轴承独立退化轨迹的提取与退化阶段的自主分割方法。该方法采用多尺度残差深度卷积自编码器对轴承自身历史监测数据进行无监督学习,实现了轴承性能退化特征的自主提取,并结合支持向量数据描述模型构建单个轴承的独立退化轨迹。引入基于去趋势化超阈波峰法的退化起始点自动检测方法与基于逻辑回归失效概率的失效阈值自主设定方法,对轴承独立退化轨迹进行了自适应分割。以轨迹分割得到的退化阶段数据为驱动,结合全阶时间幂灰色预测模型实现了轴承寿命的准确预测。试验结果表明,提出的多尺度残差深度卷积自编码器能够根据轴承各自工况环境构建反映轴承自身服役性能变化规律的退化轨迹,提出的退化轨迹自适应分割方法能够无参考地检测出轴承的起始退化点与失效阈值,从而提高轴承退化评估的科学客观性与寿命预测的工程可操作性。

基于多残差注意力深度收缩网络的超微光图像增强方法

超微光成像可在极度黑暗的环境中给观察者提供近乎白昼的视觉体验,在许多民用和军事应用中起着至关重要的作用。超微光环境下拍摄的图像和视频通常存在亮度与对比度极低、噪声水平高、场景细节和色彩严重缺失等固有缺陷,近年来,深度学习为超微光成像的研究带来了新的机遇。文中采集并提供了一组实用性更强的超微光训练数据集,提出了一种多残差注意力深度收缩网络(Multi Residual Attention Shrinkage Network),以此实现了一种新的超微光成像方法。通过成功研制的小型化样机证实了该方法的工业量产前景。实现了基于通道注意力和空间注意力的残差内注意力机制,以及基于深度软阈值收缩的外注意力机制,不仅可以有效提取并还原极低照度环境下的图像细节信息,恢复场景真实色彩,而且可以有效去除此类环境下由成像设备感光不足带来的巨量噪声。实测效果显示该方法可对极低照度环境进行有效的增强且实时性高。通过与多种业界最新方法比较,文中方法在主观视觉体验以及客观参数两方面均表现更好。

多窗谱小波阈值估计的维纳滤波语音增强算法

提出一种多窗谱小波阈值估计的维纳滤波语音增强算法.该算法采用多窗谱估计噪声功率谱,结合改进的小波阈值法减小其误差,再计算先验信噪比.改进后的阈值函数具有连续性,增加了小波系数的平滑度,实现了信噪比的平滑.与传统软、硬阈值函数算法相比,文中算法具有更高的信噪比,能较好地抑制噪声,有效提高语音清晰度.

基于多源约束自适应视觉SLAM关键帧选取研究

该文针对现有关键帧选择方法在复杂场景下的稳定性和适应性方面不足问题,提出一种多源约束的自适应视觉SLAM关键帧选取方法。该算法基于相机几何测量原理,设计自适应阈值进行关键帧选取策略;针对复杂环境下的剧烈运动情况,设计基于IMU的实时状态检测机制和熵函数约束标准,进一步提高关键帧选取的稳定性和适应性。在EuRoC数据集和TUM数据集上对该方法进行定性和定量评估。在单目惯性和立体惯性模式下,将估计轨迹与参考轨迹进行对比,以绝对轨迹误差(absolute trajectory error,ATE)、关键帧数量和运行时间作为评判指标,并与ORB-SLAM3方法进行比较。结果显示,提出的方法可显著提高视觉SLAM在复杂环境下的定位精度和稳定性。

基于增强多头注意力机制的Optuna-BiGRU测井岩性识别

测井岩性识别是油气勘探开发中至关重要的内容。针对现有算法模型在处理测井曲线数据时,无法有效捕获曲线内部深层关联和深度方向关系、拟合能力较弱、难以准确提取关键特征、噪声干扰以及模型超参数调优过程复杂困难等问题,提出了一种通过Optuna超参数优化双向门循环单元(Optuna-BiGRU)结合增强多头注意力机制(EMHA)的测井岩性识别模型——Optuna-BiGRU-EMHA模型。该模型引入残差机制和层归一化以改进多头注意力机制模块,并结合双向门循环单元(BiGRU)解决了处理测井数据时的问题,同时使用Optuna超参数优化框架和小波包自适应阈值方法分别解决了超参数调优和噪声干扰问题。首先通过交会图分析和敏感性箱线图分析选取自然伽马、深感应电阻率、中子-密度孔隙度、平均中子-密度孔隙度和岩性密度5个特征参数的测井数据,通过小波包自适应阈值方法对数据进行去噪,并将测井数据分割成数据块,然后利用Optuna框架优化BiGRU-EMHA模型超参数,最后通过实验对比K-近邻算法(KNN)、随机森林(RF)、极端梯度提升算法(XGBoost)、长短期记忆(LSTM)神经网络、BiGRU、双向长短期记忆(BiLSTM)神经网络、BiGRU-MHA、Optuna-BiGRU-EMHA等8种模型在测井岩性识别中的精度。结果表明:Optuna-BiGRU-EMHA模型识别准确率达到80%,相对于传统机器学习模型和深度学习模型,综合岩性识别准确率分别提高15.94%~23.14%和3.93%~15.94%,该模型为常规测井岩性识别提供了坚实的理论支持。

UFMC系统中基于阈值的改进DFT信道估计算法

针对通用滤波多载波(Universal Filtered Multi-Carrier,UFMC)系统中传统的基于离散傅里叶变换(Discrete Fourier Transform,DFT)的信道估计算法存在噪声区间不连续、信道长度内噪声消除不彻底、信号经DFT运算后频谱泄露高等问题,提出一种基于阈值的改进DFT信道估计算法。算法在频域通过加窗以降低信道冲激响应(Channel Impulse Response,CIR)能量泄露,在时域通过设置阈值和指数平滑可最大限度滤除样本内噪声。理论分析与仿真结果表明,与传统的DFT估计算法相比,所提算法误比特率(Bit Error Ratio,BER)更低,在高信噪比下有1.5 dB左右的性能提升,能够进一步减少噪声对系统性能的影响,可有效提高信道估计的准确度。

四向加权香农熵最大化导向的自动阈值分割方法

灰度图像的灰度直方图可以呈现出无峰、单峰、双峰或多峰的形态特征,但传统熵阈值分割方法大多仅适合处理具有单峰或双峰形态特征的灰度图像。为了提高熵阈值分割方法的分割精度和分割适应性,提出一种四向加权香农熵最大化导向的自动阈值分割方法 FWSE(Four-directional Weighted Shannon Entropy)。首先用新设计的方向性Prewitt卷积核在4个方向进行多尺度乘积变换(MPT),以获得一系列方向性MPT图像;再基于三次样条插值函数和曲率最大化准则自动计算出每个方向的最优MPT图像;其次在每个方向上通过内外轮廓图像对最优MPT图像的像素进行重新取样,以获取重构的灰度直方图,并在此基础上计算相应的香农熵;最后以4个方向的加权香农熵最大化为准则选取最佳分割阈值。与新近的3种阈值分割方法以及2种非阈值分割方法在4幅合成图像和100幅真实世界图像上进行实验,结果显示:在合成图像上,FWSE方法的平均马修斯相关系数(MCC)达到了0.999;在真实世界图像上,FWSE方法与其他5个分割方法的平均MCC分别是0.974、0.927、0.668、0.595、0.550和0.525。这表明FWSE方法具有更高的分割精度和更灵活的分割适应性。